The development of new methods for defining cardiac flow events have consistently yielded important new fields of investigation for clinical cardiology from the earliest development of phonocardiography to hemodynamics, from standard pulsed and continuous wave Doppler to new color flow mapping techniques and most recently, visualization of cardiac dynamics and cardiac flows by magnetic resonance imaging. When compared to hydrodynamics and in vitro methods for characterizing flow, most of the clinical flow imaging methods appear oversimplified, and semiquantitative at best. The dynamic color flow maps and flow images now being derived in many centers have not been currently interpretable quantitatively, nor is there a clear understanding of the determinants of flow mapping by color Doppler technologies or MRI methods. We propose the development of a staged series of flow phantoms, from characterizable steady-state simple orifices to pulsatile cardiac flow models with simulated multiple valve lesions. The study of these models under reproducible hydrodynamic conditions with highly accurate in vitro techniques of flow visualization-measurement and characterization will be undertaken, including methods of computerized quantitative particle tracing, laser Doppler anemometry, and multiple flow stream thermal labeling for separate tracing of individual streams. Once the characteristics of flow within these models have been documented in detail as it relates to laminar flow regions, flow convergence regions, high velocity jets swirling and turbulence, this knowledge will be used to guide the imaging of those techniques with digital acquisition, reconstruction and redisplay of the data, recomputation and new displays. These investigations will yield an enhanced relevant flow phenomena within the cardiovascular system. The understanding of these events should aid and accelerate the development of quantitative capabilities for all cardiac imaging modalities capable of flow visualization.